Richard Li
@rdli.bsky.social
Entrepreneur. Teacher. Data and AI. LLMs.
Pinned
Richard Li
@rdli.bsky.social
· Jan 22
7 Lessons from building an AI application
7 Lessons from building an AI application for a year
www.thelis.org
I've been building an #ai application for a little while now, and wrote up my 7 macro takeaways about building an AI app that I didn't know when I started.
www.thelis.org/blog/lessons...
www.thelis.org/blog/lessons...
#Agents aren’t the future—they’re already here.
But building them? That takes a whole new stack.
🧠 AI
⚙️ Durable execution
🧱 Frameworks
🗂 Context
🛠 Actuators
Check out the breakdown (with @jflomenb and @Wing_VC):
🔗 www.wing.vc/content/the...
But building them? That takes a whole new stack.
🧠 AI
⚙️ Durable execution
🧱 Frameworks
🗂 Context
🛠 Actuators
Check out the breakdown (with @jflomenb and @Wing_VC):
🔗 www.wing.vc/content/the...
The Agentic AI Runtime Stack | Wing Venture Capital
There’s no single architectural stack for agents, as they have multiple complex components. We see three distinct stacks for agentic AI: a model training stack, an inference stack, and an agentic runtime stack.
www.wing.vc
May 13, 2025 at 12:32 PM
#Agents aren’t the future—they’re already here.
But building them? That takes a whole new stack.
🧠 AI
⚙️ Durable execution
🧱 Frameworks
🗂 Context
🛠 Actuators
Check out the breakdown (with @jflomenb and @Wing_VC):
🔗 www.wing.vc/content/the...
But building them? That takes a whole new stack.
🧠 AI
⚙️ Durable execution
🧱 Frameworks
🗂 Context
🛠 Actuators
Check out the breakdown (with @jflomenb and @Wing_VC):
🔗 www.wing.vc/content/the...
Context Is King 👑. Smarter agents need richer context—not just prompts. They aggregate context from multiple sources:
🧠 Knowledge (vector databases)
💾 Memory (short/long-term)
🌐 Actuators (APIs, sensors)
Check out the full Agentic Runtime Stack:
www.wing.vc/content/the...
🧠 Knowledge (vector databases)
💾 Memory (short/long-term)
🌐 Actuators (APIs, sensors)
Check out the full Agentic Runtime Stack:
www.wing.vc/content/the...
The Agentic AI Runtime Stack | Wing Venture Capital
There’s no single architectural stack for agents, as they have multiple complex components. We see three distinct stacks for agentic AI: a model training stack, an inference stack, and an agentic runtime stack.
www.wing.vc
May 12, 2025 at 6:33 PM
Context Is King 👑. Smarter agents need richer context—not just prompts. They aggregate context from multiple sources:
🧠 Knowledge (vector databases)
💾 Memory (short/long-term)
🌐 Actuators (APIs, sensors)
Check out the full Agentic Runtime Stack:
www.wing.vc/content/the...
🧠 Knowledge (vector databases)
💾 Memory (short/long-term)
🌐 Actuators (APIs, sensors)
Check out the full Agentic Runtime Stack:
www.wing.vc/content/the...
Most apps act fast. Agents don’t.
They pause, retry, wait hours or days. That’s why they need durable execution—resilient workflows that persist across failures.
Check out our post on the agentic runtime stack:
www.wing.vc/content/the...
They pause, retry, wait hours or days. That’s why they need durable execution—resilient workflows that persist across failures.
Check out our post on the agentic runtime stack:
www.wing.vc/content/the...
The Agentic AI Runtime Stack | Wing Venture Capital
There’s no single architectural stack for agents, as they have multiple complex components. We see three distinct stacks for agentic AI: a model training stack, an inference stack, and an agentic runtime stack.
www.wing.vc
April 29, 2025 at 4:10 PM
Most apps act fast. Agents don’t.
They pause, retry, wait hours or days. That’s why they need durable execution—resilient workflows that persist across failures.
Check out our post on the agentic runtime stack:
www.wing.vc/content/the...
They pause, retry, wait hours or days. That’s why they need durable execution—resilient workflows that persist across failures.
Check out our post on the agentic runtime stack:
www.wing.vc/content/the...
Agentic frameworks are expanding fast—most started with some sort of syntactic sugar around prompts, but they've since expanded into durable execution & memory.
I teamed up with @jflomenb + @wing_vc to map the stack 👇
🔗 www.wing.vc/content/the...
#LLMops #AgenticAI
I teamed up with @jflomenb + @wing_vc to map the stack 👇
🔗 www.wing.vc/content/the...
#LLMops #AgenticAI
The Agentic AI Runtime Stack | Wing Venture Capital
There’s no single architectural stack for agents, as they have multiple complex components. We see three distinct stacks for agentic AI: a model training stack, an inference stack, and an agentic runtime stack.
www.wing.vc
April 28, 2025 at 11:14 AM
Agentic frameworks are expanding fast—most started with some sort of syntactic sugar around prompts, but they've since expanded into durable execution & memory.
I teamed up with @jflomenb + @wing_vc to map the stack 👇
🔗 www.wing.vc/content/the...
#LLMops #AgenticAI
I teamed up with @jflomenb + @wing_vc to map the stack 👇
🔗 www.wing.vc/content/the...
#LLMops #AgenticAI
🚀 Agents aren't the future—they're here.
From legal #AI to AI software engineers, #AgenticAI systems are reshaping how work gets done. Building them requires a new stack: the Agentic Runtime Stack.
🧠 AI
⚙️ Durable Execution
🧱 Agentic Frameworks
🗂 Context
🛠 Actuators
www.wing.vc/content/the-...
From legal #AI to AI software engineers, #AgenticAI systems are reshaping how work gets done. Building them requires a new stack: the Agentic Runtime Stack.
🧠 AI
⚙️ Durable Execution
🧱 Agentic Frameworks
🗂 Context
🛠 Actuators
www.wing.vc/content/the-...
The Agentic AI Runtime Stack | Wing Venture Capital
There’s no single architectural stack for agents, as they have multiple complex components. We see three distinct stacks for agentic AI: a model training stack, an inference stack, and an agentic runt...
www.wing.vc
April 25, 2025 at 5:08 PM
🚀 Agents aren't the future—they're here.
From legal #AI to AI software engineers, #AgenticAI systems are reshaping how work gets done. Building them requires a new stack: the Agentic Runtime Stack.
🧠 AI
⚙️ Durable Execution
🧱 Agentic Frameworks
🗂 Context
🛠 Actuators
www.wing.vc/content/the-...
From legal #AI to AI software engineers, #AgenticAI systems are reshaping how work gets done. Building them requires a new stack: the Agentic Runtime Stack.
🧠 AI
⚙️ Durable Execution
🧱 Agentic Frameworks
🗂 Context
🛠 Actuators
www.wing.vc/content/the-...
Reasoning LLMs like OpenAI o1 and DeepSeek-R1 have shown how #AI can tackle complex problems by breaking them down. I've been doing work in this area, and have written up my notes here: www.thelis.org/blog/reasoni...
Engineering reasoning LLMs: Notes and Observations
Engineering reasoning LLMs: Notes and Observations
www.thelis.org
March 26, 2025 at 6:25 PM
Reasoning LLMs like OpenAI o1 and DeepSeek-R1 have shown how #AI can tackle complex problems by breaking them down. I've been doing work in this area, and have written up my notes here: www.thelis.org/blog/reasoni...
Reposted by Richard Li
The era of #agenticAI is here.
#AIagents are replacing manual workflows—making decisions, taking action, and scaling fast.
Learn how to design and implement the next generation of AI-powered services.
Register now: register.gotowebinar.com/register/907...
#Java #AI @infoq.com @rdli.bsky.social
#AIagents are replacing manual workflows—making decisions, taking action, and scaling fast.
Learn how to design and implement the next generation of AI-powered services.
Register now: register.gotowebinar.com/register/907...
#Java #AI @infoq.com @rdli.bsky.social
March 24, 2025 at 8:12 PM
The era of #agenticAI is here.
#AIagents are replacing manual workflows—making decisions, taking action, and scaling fast.
Learn how to design and implement the next generation of AI-powered services.
Register now: register.gotowebinar.com/register/907...
#Java #AI @infoq.com @rdli.bsky.social
#AIagents are replacing manual workflows—making decisions, taking action, and scaling fast.
Learn how to design and implement the next generation of AI-powered services.
Register now: register.gotowebinar.com/register/907...
#Java #AI @infoq.com @rdli.bsky.social
Setting a price that you're willing to pay is a critical component of software product requirements -- but many PMs don't communicate the price. This seemingly simple bit of information enables the entire org to make better tradeoffs during the software process.
www.thelis.org/blog/set-a-p...
www.thelis.org/blog/set-a-p...
The missing piece of software product requirements: Price
product requirements should always have a price
www.thelis.org
February 21, 2025 at 2:46 PM
Setting a price that you're willing to pay is a critical component of software product requirements -- but many PMs don't communicate the price. This seemingly simple bit of information enables the entire org to make better tradeoffs during the software process.
www.thelis.org/blog/set-a-p...
www.thelis.org/blog/set-a-p...
The #AI market is 🔥. Anthropic is raising another $2B at a $60B valuation when their run-rate revenue at the end of 2024 was $1B. I ran some back-of-the-envelope math to figure out under what scenarios Anthropic stock options are valuable.
www.thelis.org/blog/anthrop...
www.thelis.org/blog/anthrop...
February 5, 2025 at 1:35 PM
The #AI market is 🔥. Anthropic is raising another $2B at a $60B valuation when their run-rate revenue at the end of 2024 was $1B. I ran some back-of-the-envelope math to figure out under what scenarios Anthropic stock options are valuable.
www.thelis.org/blog/anthrop...
www.thelis.org/blog/anthrop...
@daringfireball.bsky.social on less than a year ago, you said "While I’m being grumpy, I’ll even take issue with the notion ... that Apple is “behind in AI”? daringfireball.net/linked/2024/.... Do you still agree?
Fitting Facts to the Narrative at The Washington Post
Link to: https://www.washingtonpost.com/technology/2024/05/02/apple-sales-ai-iphone-earnings-regulations/
daringfireball.net
January 31, 2025 at 2:21 PM
@daringfireball.bsky.social on less than a year ago, you said "While I’m being grumpy, I’ll even take issue with the notion ... that Apple is “behind in AI”? daringfireball.net/linked/2024/.... Do you still agree?
A common response I got to my post was "Try $OTHER_AI_FRAMEWORK." I wasn't trying to critique @langchain.bsky.social or @llamaindex.bsky.social. AI dev frameworks in general are solving the wrong problem, & the real challenges with #ai apps are elsewhere:
www.thelis.org/blog/ai-dev-...
www.thelis.org/blog/ai-dev-...
January 29, 2025 at 5:47 PM
A common response I got to my post was "Try $OTHER_AI_FRAMEWORK." I wasn't trying to critique @langchain.bsky.social or @llamaindex.bsky.social. AI dev frameworks in general are solving the wrong problem, & the real challenges with #ai apps are elsewhere:
www.thelis.org/blog/ai-dev-...
www.thelis.org/blog/ai-dev-...
I've been building an #ai application for a little while now, and wrote up my 7 macro takeaways about building an AI app that I didn't know when I started.
www.thelis.org/blog/lessons...
www.thelis.org/blog/lessons...
7 Lessons from building an AI application
7 Lessons from building an AI application for a year
www.thelis.org
January 22, 2025 at 7:15 PM
I've been building an #ai application for a little while now, and wrote up my 7 macro takeaways about building an AI app that I didn't know when I started.
www.thelis.org/blog/lessons...
www.thelis.org/blog/lessons...
Reposted by Richard Li
Rain Saturday afternoon, break Sunday morning. Snow/mix arrives Sunday afternoon. Storm ends Monday morning. Still 48 hours or so until snow underway. Expect the bands to shift as we get closer to storm. Cold air comes in before so fluffy snow can pile up faster. #Snow
January 17, 2025 at 11:17 PM
Rain Saturday afternoon, break Sunday morning. Snow/mix arrives Sunday afternoon. Storm ends Monday morning. Still 48 hours or so until snow underway. Expect the bands to shift as we get closer to storm. Cold air comes in before so fluffy snow can pile up faster. #Snow
The vendor email that wishes me “Happy Holidays” or “Happy New Year” is … just irritating. Don’t send it.
January 1, 2025 at 6:43 PM
The vendor email that wishes me “Happy Holidays” or “Happy New Year” is … just irritating. Don’t send it.
Reposted by Richard Li
THIS IS HUGE! Researchers at TAU and the University of Lisbon have developed a groundbreaking nano-vaccine against COVID-19, delivered via nasal spray WITHOUT requiring cold storage. Preclincial studies show the vaccine was effective against ALL major SARS-CoV-2 variants. Let’s talk about that! 🧪🧵⬇️
December 9, 2024 at 5:00 PM
THIS IS HUGE! Researchers at TAU and the University of Lisbon have developed a groundbreaking nano-vaccine against COVID-19, delivered via nasal spray WITHOUT requiring cold storage. Preclincial studies show the vaccine was effective against ALL major SARS-CoV-2 variants. Let’s talk about that! 🧪🧵⬇️
This is 💯
Just realized I could add accounts to lists without also following them.
Great for my breaking news list. I can include everything and not have some of the accounts I was unsure about filling up my primary feed !
Great for my breaking news list. I can include everything and not have some of the accounts I was unsure about filling up my primary feed !
December 1, 2024 at 2:16 PM
This is 💯
I woke up one morning on blue sky and discovered a swarm of cloud native refugees.
November 20, 2024 at 11:37 PM
I woke up one morning on blue sky and discovered a swarm of cloud native refugees.
Reposted by Richard Li
I attended a meetup in Seattle and got to spend time with @bnewbold.net and @pfrazee.com from the Bluesky engineering team.
I walked away with an understanding of the role DIDs and PDSs play in the grand scheme of things, and why the most exciting work sits higher up the stack. I almost unretired.
I walked away with an understanding of the role DIDs and PDSs play in the grand scheme of things, and why the most exciting work sits higher up the stack. I almost unretired.
November 20, 2024 at 6:22 PM
I attended a meetup in Seattle and got to spend time with @bnewbold.net and @pfrazee.com from the Bluesky engineering team.
I walked away with an understanding of the role DIDs and PDSs play in the grand scheme of things, and why the most exciting work sits higher up the stack. I almost unretired.
I walked away with an understanding of the role DIDs and PDSs play in the grand scheme of things, and why the most exciting work sits higher up the stack. I almost unretired.
Ranking worst to best hyperscaler cloud IAM.
3. Azure (worst)
2. Amazon (less worse)
1. Google (least worse)
I know RBAC is complicated and there’s legacy but why can’t the documentation be straightforward at least?
3. Azure (worst)
2. Amazon (less worse)
1. Google (least worse)
I know RBAC is complicated and there’s legacy but why can’t the documentation be straightforward at least?
November 12, 2024 at 11:59 AM
Ranking worst to best hyperscaler cloud IAM.
3. Azure (worst)
2. Amazon (less worse)
1. Google (least worse)
I know RBAC is complicated and there’s legacy but why can’t the documentation be straightforward at least?
3. Azure (worst)
2. Amazon (less worse)
1. Google (least worse)
I know RBAC is complicated and there’s legacy but why can’t the documentation be straightforward at least?
Reposted by Richard Li
I'm watching some folks reverse engineer the xz backdoor, sharing some *preliminary* analysis with permission.
The hooked RSA_public_decrypt verifies a signature on the server's host key by a fixed Ed448 key, and then passes a payload to system().
It's RCE, not auth bypass, and gated/unreplayable.
The hooked RSA_public_decrypt verifies a signature on the server's host key by a fixed Ed448 key, and then passes a payload to system().
It's RCE, not auth bypass, and gated/unreplayable.
This might be the best executed supply chain attack we've seen described in the open, and it's a nightmare scenario: malicious, competent, authorized upstream in a widely used library.
Looks like this got caught by chance. Wonder how long it would have taken otherwise.
Looks like this got caught by chance. Wonder how long it would have taken otherwise.
Woah. Backdoor in liblzma targeting ssh servers.
www.openwall.com/lists/oss-se...
It has everything: malicious upstream, masterful obfuscation, detection due to performance degradation, inclusion in OpenSSH via distro patches for systemd support…
Now I’m curious what it does in RSA_public_decrypt
www.openwall.com/lists/oss-se...
It has everything: malicious upstream, masterful obfuscation, detection due to performance degradation, inclusion in OpenSSH via distro patches for systemd support…
Now I’m curious what it does in RSA_public_decrypt
March 30, 2024 at 5:13 PM
I'm watching some folks reverse engineer the xz backdoor, sharing some *preliminary* analysis with permission.
The hooked RSA_public_decrypt verifies a signature on the server's host key by a fixed Ed448 key, and then passes a payload to system().
It's RCE, not auth bypass, and gated/unreplayable.
The hooked RSA_public_decrypt verifies a signature on the server's host key by a fixed Ed448 key, and then passes a payload to system().
It's RCE, not auth bypass, and gated/unreplayable.
I’ve been thinking a lot about how to analyze data (dashboards can’t be the right answer!) and wrote up some thoughts here: www.amorphousdata.com/blog/data-is.... #data
Toggle Menu
www.amorphousdata.com
January 9, 2024 at 3:54 PM
I’ve been thinking a lot about how to analyze data (dashboards can’t be the right answer!) and wrote up some thoughts here: www.amorphousdata.com/blog/data-is.... #data
Any #data or #deeplearning people on here for me to follow?
December 23, 2023 at 12:24 PM
Any #data or #deeplearning people on here for me to follow?